Project Summary Up to one third of pediatric patients with epilepsy are medically intractable and require resective neurosurgery to gain seizure freedom. If performed early in life, epilepsy surgery can offer significant cognitive and developmental gains improving the quality of life in children with epilepsy. Crucial to the success of surgery is the availability of a robust presurgical biomarker to identify the epileptogenic zone (EZ), the area of the cortex that is indispensable for the generation of epileptic seizures. Complete resection of the EZ may lead to seizure freedom and may improve social, psychological and cognitive development. Since the EZ cannot be measured directly, its location is estimated based on concordant data from a multitude of tests. Yet, the results of these tests are often insufficiently concordant or inconclusive, leading to the need for intracranial long-term recordings, which present many disadvantages due to their invasiveness. Thus, the presurgical evaluation of patient with refractory epilepsy is often challenging or unsuccessful. To improve the efficacy of surgery, there is an overriding need for a reliable noninvasive biomarker of the EZ. High Frequency Oscillations (HFOs) have established over the last decade as biomarkers for the delineation of the EZ. However, their effective use as biomarkers of the EZ is still at an early stage. One of the primary challenges remains the difficulty to detect and localize them noninvasively. This R21 application aims to noninvasively detect and reliably localize HFOs from pediatric patients with refractory epilepsy using high-density scalp electroencephalography (EEG) and magnetoencephalography (MEG), identify their onset generator, and correlate the resection of this generator with patients? postsurgical outcome. We hypothesize that high-density scalp EEG and MEG can noninvasively identify and accurately localize the HFO onset generator whose removal leads to a better postsurgical outcome compared to the removal of the HFO propagation area. To test our hypothesis, we specifically aim to: (i) detect and localize noninvasively interictal HFOs with EEG and MEG, (ii) localize the onset generator of interictal HFOs and correlate it with the SOZ, and (iii) correlate the resection ratio of the HFO onset generator with patient?s outcome. We will pursue our aims in children (0-18 years old) with refractory epilepsy. We will also record EEG and MEG data from 10 age-matched typically-developing children, and will use these control data to distinguish pathological from physiological HFOs. For young children, we will use the BabyMEG, a pediatric MEG system with unique features designed specifically for children up to 3 years old. Our research will have a direct impact on the quality of life of children with epilepsy because it will provide a new noninvasive biomarker for the precise delineation of EZ that will eventually: (i) improve the success rate of epilepsy surgery; (ii) reduce postsurgical neurological deficits; (iii) limit long-term invasive recordings; (iv) enable to assess the efficacy of therapeutic interventions without waiting for a seizure to occur; and (v) permit definitive differential diagnosis of epilepsy from acute symptomatic seizures so treatment can begin immediately.